Realistic Multi-Step Counterfactual Explanations for Learning-Based 2D LiDAR Robot Control
IEEE Access · 2026 · 14 · 113623-113643
DOI: 10.1109/access.2026.3715401 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Sindre B. Remman und 1 weiteren Autor:innen. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Research Council of Norway, Explainable AI systems for gradual industry adoption (EXAIGON) Project.
Abstract (Originalsprache)
In this paper, we propose a novel method for generating realistic multi-step counterfactual explanations for machine learning-controlled robots using 2D Light Detection and Ranging (LiDAR) data. Machine learning algorithms based on black-box models such as neural networks suffer from a lack of interpretability. This lack of interpretability is critical when such models are used to control robots, which are often deployed in safety-critical environments. Robotic policies also act in a closed loop with the environment and rely on high-dimensional sensor input, which makes their behavior difficult to explain from a single observation. Our method extends our previous single-step approach by generating explanations based on forward simulations of the robot’s dynamics. The proposed approach explains the behavior of agents over a time horizon, making it suitable for reinforcement learning (RL)-based policies. To support these explanations, we introduce new cost functions that enable users to pose trajectory-level questions such as “why does the robot not visit this area instead?”. We evaluate our method on a TurtleBot3 platform controlled by an RL policy trained to navigate in unknown environments. We compare the proposed multi-step method with the single-step approach and show that multi-step counterfactuals provide more informative explanations. In particular, the method shows how alternative obstacle configurations could affect the robot’s trajectory over several time steps, revealing behavior that is missed by single-step explanations.
Förderung
- Research Council of Norway
- Explainable AI systems for gradual industry adoption (EXAIGON) Project
Zitieren
APA
Remman, S. B., & Lekkas, A. M. (2026). Realistic Multi-Step Counterfactual Explanations for Learning-Based 2D LiDAR Robot Control. IEEE Access, 14, 113623-113643. https://doi.org/10.1109/access.2026.3715401
Vancouver
Remman SB, Lekkas AM. Realistic Multi-Step Counterfactual Explanations for Learning-Based 2D LiDAR Robot Control. IEEE Access. 2026;14:113623-113643. doi:10.1109/access.2026.3715401
BibTeX
@article{Remman2026,
author = {Remman, Sindre B. and Lekkas, Anastasios M.},
title = {Realistic Multi-Step Counterfactual Explanations for Learning-Based 2D LiDAR Robot Control},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {113623--113643},
doi = {10.1109/access.2026.3715401}
}
EUVIMED